Nonalcoholic fatty liver disease (NAFLD) and nonalcoholic. Noninvasive Assessment of Nonalcoholic Fatty Liver Disease in Obese or Overweight Patients

Transcription

1 CLINICAL GASTROENTEROLOGY AND HEPATOLOGY 2012;10: Noninvasive Assessment of Nonalcoholic Fatty Liver Disease in Obese or Overweight Patients SVEN M. A. FRANCQUE,* AN VERRIJKEN, ILSE MERTENS, GUY HUBENS, ERIC VAN MARCK, PAUL PELCKMANS,* PETER MICHIELSEN,* and LUC VAN GAAL *Department of Gastroenterology and Hepatology, Department of Endocrinology, Diabetes, and Metabolism, Department of Abdominal Surgery, and Department of Pathology, Antwerp University Hospital, Antwerp, Belgium This article has an accompanying continuing medical education activity on page e87. Learning Objective At the end of this activity, the successful learner will be able to select and interpret the parameters they need to assess the presence of NASH in obese patients without the need for a liver biopsy. BACKGROUND & AIMS: Reliable noninvasive tools are needed for staging nonalcoholic fatty liver disease (NAFLD). Published scoring systems have not been validated in prospective assessments of unselected patients. We aimed to identify factors that predicted development of nonalcoholic steatohepatitis (NASH) in a large group of overweight or obese patients and compared these with established factors. METHODS: We performed a prospective analysis of factors associated with the development and severity of NAFLD in patients at a single obesity center. We evaluated liver involvement in 542 patients by a large set of routine and non-routine parameters, including ultrasound and genetic testing. Those suspected of having NA- FLD underwent liver biopsy (57.7%). Patients were divided into design cohort (n 200) and validation cohort (n 113) to identify factors associated with the presence and severity of NAFLD and NASH. RESULTS: Factors independently associated with development of NASH included increased levels of alanine aminotransferase (ALT), fasting levels of C-peptide, and ultrasound steatosis scores (USSs), with area under the receiver operating curve (AUROC) values of in the design cohort and in the validation cohort. NASH activity scores also correlated with level of ALT, USS, and fasting level of C-peptide (R ). Independent predictors of advanced fibrosis included waist circumference and level of aspartate aminotransferase (AUROC values of and for design and validation cohorts, respectively; negative predictive values of 98% and 97%, respectively, for a cutoff of 2.14). Previously published scoring systems had significantly lower AUROC values. Levels of CK18 and PNPLA3 polymorphisms correlated with development of NASH but did not add value. CONCLU- SIONS: Parameters routinely analyzed in assessing obese patients can be used to determine the presence of NASH and advanced fibrosis. Non-routine tests do not increase diagnostic accuracy. Previously published scores are significantly less accurate. Keywords: Risk Factor; Liver Disease; Complication; Genetic Test; Liver Stiffness. Nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH) are increasingly recognized as important causes of liver-related and non liver-related morbidity and mortality, with an increasing impact on health care resources. 1,2 The pathophysiology of NAFLD and its related morbidities remain largely unknown. Merely on the basis of epidemiologic data, NAFLD and NASH seem to be intimately linked to (visceral) obesity, insulin resistance and diabetes, and metabolic syndrome (MS). 2 Many people are hence at risk for NAFLD and NAFLD-related complications. The gold standard for the diagnosis of NAFLD and NASH is liver histology. Because of its inherent risks and for logistical reasons, liver biopsy cannot be applied on a large scale. Therefore, several attempts have been made to develop noninvasive tools for NAFLD and NASH diagnosis. 3 9 Because those tools are mostly designed in highly selected populations and/or are based on retrospective analysis, their diagnostic accuracy is frequently unsatisfactory when applied to randomly presenting patients. 10 In the present study, we prospectively assessed the presence and severity of NAFLD in a population at risk by a two-step approach to avoid some identifiable mechanisms of selection bias. The patients were subsequently divided into a design cohort and a validation cohort to develop scores that must allow one to accurately and noninvasively identify the presence and severity of NAFLD and NASH in randomly presenting patients. Methods Metabolic Work-up Patients presenting at the obesity clinic for a problem of overweight underwent a series of examinations including a metabolic and a liver-specific program as previously reported 11 (Supplementary Material). Abbreviations used in this paper: ABT, aminopyrine breath test; ALT, alanine aminotransferase; APRI, AST-to-platelet index; AST, aspartate aminotransferase; CK-18, cytokeratin 18; MS, metabolic syndrome; NAFLD, nonalcoholic fatty liver disease; NAS, NASH activity score; NASH, nonalcoholic steatohepatitis; PNPLA3, patatin-like phospholipase domain-containing protein 3; USS, ultrasound steatosis score by the AGA Institute /$

2 October 2012 PROSPECTIVE ASSESSMENT OF NAFLD IN OBESITY 1163 Because liver steatosis seems to be closely related to MS, 1 MS features were specifically recorded according to both the US Third Adult Treatment Panel of the National Cholesterol Education Program 12 and the International Diabetes Federation. 13 Hepatologic Work-up The liver-specific program, approved by the Ethics Committee of the Antwerp University Hospital, included additional blood analysis (Supplementary Material), an abdominal Doppler ultrasound, and an aminopyrine breath test (ABT) as a measure for liver metabolic reserve. 14 The ultrasound appearance of the liver parenchyma was scored by using a modification of the Saverymuttu classification 15 by making the sum of the echogenicity of the liver parenchyma compared with the renal parenchyma (no hyperechogenicity, 0; mild-to-moderate hyperechogenicity, 1; moderate-to-severe hyperechogenicity, 2) and posterior beam attenuation (absent, 0; present, 1), resulting in an ultrasound steatosis score (USS) ranging from 0 to 3. Patients were excluded from further analysis in case of significant alcohol consumption ( 20 g/d), history of bariatric surgery, diagnosis of another liver disease, or preexisting diabetes. The possibility of liver involvement was defined by abnormal liver tests (aspartate aminotransferase [AST] and/or alanine aminotransferase [ALT] and/or -glutamyl transpeptidase and/or alkaline phosphatase); liver ultrasound abnormality (steatotic liver [USS 1) 15 ; and abnormal ABT. 14 For ALT, 3 different cutoff levels were used: the upper limit of normal set by the biochemistry laboratory (56 IU/L), the classic cutoff of 40 IU/L, and the limits proposed by Prati et al 16 (30 IU/L in men, 19 IU/L in women). Liver Biopsy When one or more of these criteria were met, a liver biopsy was proposed. For patients going to surgery, a liver biopsy was proposed regardless of the preset criteria. Liver biopsy was performed (after additional informed consent) percutaneously (16-gauge Menghini) or perioperatively (14-gauge Tru-Cut). Hematoxylin-eosin stain, Sirius red stain, reticulin stain, and Perl s iron stain were routinely performed on all biopsies, which were reanalyzed by one experienced pathologist (E.V.M.) who was blinded for any clinical data. The different histologic features of NAFLD were assessed by using the NASH Clinical Research Network Scoring System. 17 The NASH Activity Score (NAS) was calculated by making the sum of the scores for steatosis, lobular inflammation, and ballooning. 17 Significant fibrosis was defined as a fibrosis score 2. Advanced fibrosis was defined as a fibrosis score 3. Since the publication of the NASH Clinical Research Network Scoring System, the definition of NASH by a NAS 5 has been widely adopted. This definition is, however, slightly different from the former definition of Brunt et al, 18 and its use outside the setting of interventional studies was recently questioned. 19 We therefore used both definitions separately. Additional Tests Cytokeratin 18 (CK-18) has shown promising properties as a serum marker of liver fibrosis and was determined by using a two enzyme-linked immunosorbent assay (PEVIVA AB, Bromma, Sweden) according to the manufacturer s instructions. 20 Single nucleotide polymorphisms in the patatin-like phospholipase domain-containing protein 3 (PNPLA3) locus on chromosome 22 have recently been associated with NASH independently from metabolic factors. 21 We studied the rs polymorphism, which seems to be the most important single nucleotide polymorphism. Scoring Systems To design and validate scores that might predict the presence and severity of NAFLD and fibrosis, patients with liver biopsy were randomly assigned to a design cohort (n 200) or a validation cohort (n 113) by the SPSS 18.0 (SPSS Inc, Chicago, IL) software package after all data had been obtained. The following published scores were used for comparison: NAFLD liver fat score, 3 the fatty liver index, 4 the body mass index, AST/ALT ratio, and diabetes mellitus score, 5 the AST-toplatelet index (APRI), 6 the fibrosis test based on age, AST, ALT, and platelet count index, 7 the NAFLD fibrosis score, 8 and Forns index 9 (Supplementary Material). Statistical Analysis See Supplementary Material. Results Main Characteristics Between October 2005 and October 2010, 553 patients were screened. Eleven patients were excluded from further analysis because of the discovery of a formerly unknown chronic liver disease (1 primary biliary cirrhosis, 2 hemochromatosis, 1 chronic hepatitis C, 2 chronic hepatitis B, 1 autoimmune hepatitis) or because of alcohol consumption disclosed on repeated interrogation (4 patients). The main characteristics of the remaining 542 patients are listed in Table 1. Histology In 64 of 542 patients (11.8%), none of the criteria to propose a biopsy were present. In the other patients, at least one criterion was met. Liver biopsy was performed in 313 of 478 of those patients (65.5%, or 57.7% of the overall cohort of 542 patients). Liver biopsy was performed during bariatric surgery in 133 of 313 (42.5%). Mean biopsy length was mm (range, 8 45), and the mean number of portal tracts was (range, 5 25). The distributions according to the grade of steatosis (Figure 1A), the NAS (Figure 1B), and the stage of fibrosis (Figure 1C) are shown in Figure 1. On the basis of the definition by Brunt et al, of 313 (52.1%) had NASH. On the basis of the definition by Kleiner et al, of 313 (33.5%) had NASH. Three patients had NASH on the basis of Kleiner et al but not by Brunt et al, and 61 patients had NASH according to Brunt et al, but a NAS 5. The latter 61 all had a borderline or possible NASH (NAS 3 or 4) according to Kleiner et al. Design and Validation Cohorts Besides a higher mean arterial blood pressure in the design cohort, no significant differences were noted (Supplementary Material).

3 1164 FRANCQUE ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 10 Table 1. Main Characteristics of the Overall Patient Cohort (n 542) Units Range Sex, M/F %/% 28.6/71.4 Age y BMI kg/m Waist cm WHR VAT cm MABP mm Hg LDH U/L AST U/L ALT U/L AST/ALT GGT U/L Total cholesterol mg/dl HDL cholesterol mg/dl TG mg/dl Fasting glucose mg/dl Fasting insulin U/mL Fasting C-peptide nmol/l HBA1c % QUICKI index HOMA IR USS Diabetes absent/present %/% 91.6/8.4 Liver right craniocaudal diameter mm Spleen diameter mm ABT peak excretion % ABT cumulative excretion % Number criteria MS NCEP ATP III Number criteria MS IDF MS NCEP ATP III absent/present %/% 49.3/50.7 MS IDF absent/present %/% 47.9/52.1 PNPLA3 polymorphism (CC/CG/GG) n (%) 324 (60.7)/189 (34.9)/29 (5.4) CK-18 U/L NOTE. Continuous variables are presented as mean standard deviation. BMI, body mass index; GGT, -glutamyl transpeptidase; HBA1c, glycosylated hemoglobin; HDL, high-density lipoprotein; HOMA IR, homeostasis model of assessment insulin resistance; IDF, International Diabetes Federation; LDH, lactate dehydrogenase; MABP, mean arterial blood pressure; NCEP ATP III, US Third Adult Treatment Panel of the National Cholesterol Education Program; QUICKI, quantitative assessment check index; TG, triglycerides; VAT, visceral adipose tissue; WHR, waist-to-hip ratio; y/n, yes or no. Features of Nonalcoholic Steatohepatitis: Steatosis, Ballooning, Lobular Inflammation, and Fibrosis Nonalcoholic Steatohepatitis According to the Definition by Brunt et al The factors that were significantly related to the presence of NASH as defined by Brunt et al 18 in a univariate binary logistic regression analysis in the design cohort are listed in Supplementary Table 4. In a multivariate analysis, only ALT 40 U/L, the USS, and fasting C-peptide appeared to be independent predictors of the presence of NASH (Supplementary Table 4). The Antwerp NASH score 1 to predict the diagnosis of NASH according to the definition by Brunt et al 18 had the following formula: [1.315 (ALT 40 U/L: no 0; yes 1)] [1.084 USS] [1.223 fasting C-peptide (nmol/l)]. This score had an R The areas under the receiver operating curve of this score and the reference scores are shown in Table 2 (design and validation cohorts) and Figure 2A (design cohort). The other diagnostic accuracy indexes are listed in Table 3. Nonalcoholic Steatohepatitis According to the Definition by Kleiner et al The Nonalcoholic Steatohepatitis Activity Score The factors that are significantly correlated with the severity of NASH as expressed by the NAS in a univariate analysis in the design cohort are listed in Supplementary Table 6. In a multivariate analysis ALT, the USS, and fasting C-peptide appeared to independently predict the NAS (Supplementary Table 6). The Antwerp NASH severity score to calculate the NAS had the following formula: [0.021 ALT (U/L)] [1.200

4 October 2012 PROSPECTIVE ASSESSMENT OF NAFLD IN OBESITY 1165 Figure 1. Histologic lesions. Pie plot of the distribution of patients according to the grade of steatosis (A), NAS (B), and fibrosis stage (C). USS] [0.936 fasting C-peptide (nmol/l)] (Supplementary Figure 1). This score had an R The correlation of the Antwerp NASH severity score with the NAS was (Spearman rank correlation coefficient) and in the index and validation cohorts, respectively. Significant Fibrosis The factors that were significantly related to the presence of significant fibrosis (19.5% of patients) in a univariate analysis in the design cohort are listed in Supplementary Table 7. In a multivariate analysis only AST, fasting C-peptide, and waist appeared to be independent predictors of the presence of significant fibrosis (Supplementary Table 7). The Antwerp NAFLD significant fibrosis score to predict the diagnosis of significant fibrosis had the following formula: [0.048 waist (cm)] [1.373 fasting C-peptide Table 2. AUROCs of Several Scoring Systems for the Diagnosis of NASH Defined According to Brunt et al 18 and Kleiner et al 17 for the Design and Validation Cohorts NASH according to Brunt et al NASH according to Kleiner et al AUROC 1 AUROC 2 AUROC 1 AUROC 2 Antwerp NASH score Antwerp NASH score NAFLD liver fat score FLI BARD score APRI FIB-4 index NAFLD fibrosis score Forns index AUROC1, area under the receiver operating curve of the design cohort; AUROC2, area under the receiver operating curve of the validation cohort; BARD, body mass index, AST/ALT ratio, and diabetes mellitus; FIB-4, fibrosis test based on age, AST, ALT, and platelet count; FLI, fatty liver index. (nmol/l)] [0.043 AST (U/L)]. This score had an R The areas under the receiver operating curve of this score and the reference scores are shown in Table 4 (design and validation cohorts) and Figure 2C (design cohort). The other diagnostic accuracy indexes are listed in Table 3. Advanced Fibrosis Application of the Scores to the Overall Cohort Discussion Well-documented data on NAFLD and NASH prevalences in populations at risk are rather scarce, and noninvasive tools to assess NAFLD on a large scale are warranted. 1,10 In our study, prospective assessment of liver disease in a large cohort of patients with obesity shows NASH in 52% and significant fibrosis in 19.5%. We could design and validate scoring systems for NASH and fibrosis on the basis of routine parameters. Ultrasound significantly adds to the diagnostic accuracy for NASH diagnosis, whereas genetic testing and non-routine biological tests do not have an added value. Published scores are significantly less accurate. Studies on prevalence and risk factors for NAFLD frequently suffer from methodological drawbacks. Most of the scores are based on series of biopsy-proven NAFLD patients, 3 9 who are mostly highly selected (eg, referred because of elevated aminotransferase levels or included in clinical trials), resulting in series with a relatively high prevalence of more advanced disease. Scores designed in such populations are at risk of lower accuracy when applied to populations with lower prevalences. The two-step approach of our study aimed at overcoming some of these methodological issues. In a first step, all patients presenting to the obesity clinic were prospectively assessed without a priori suspicion of liver

5 1166 FRANCQUE ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 10 Figure 2. Comparison of scoring systems. ROC of the different scoring systems for the diagnosis of NASH defined according to Brunt et al 18 (A), for the diagnosis of NASH defined according to Kleiner et al 17 (B), for the diagnosis of significant fibrosis (C), and for the diagnosis of advanced fibrosis (D) in the design cohort. disease. If there was any indication of NAFLD, a biopsy was proposed, representing the second step. We believe that this two-step prospective approach avoided some important selection bias. Furthermore, the results of the design cohort could be closely reproduced in the validation cohort, supporting their general applicability. We therefore estimate that the scores can reliably be used to prospectively screen patients who are overweight. Further studies are warranted to test the diagnostic accuracy in nonobese patients or in other patients at risk (especially diabetes). The natural history and clinical relevance of NAFLD and NASH are still largely unknown. 10 NAFLD has been shown to be independently associated with increased overall, liver-related, and cardiovascular mortality. 10 Although the liver-related (but not cardiovascular) mortality is higher in patients with NASH compared with simple steatosis (suggesting that pro- Table 3. Diagnostic Accuracy Indices Score Cutoff Sens 1 Sens 2 Spec 1 Spec 2 PPV 1 PPV 2 NPV 1 NPV 2 Antwerp NASH score Antwerp NASH score Antwerp NAFLD significant fibrosis score Antwerp NAFLD advanced fibrosis score NOTE. Sensitivity (Sens), specificity (Spec), positive predictive value (PPV), and negative predictive value (NPV) for the different Antwerp scores in the design cohort ( 1 ) and validation cohort ( 2 ) are listed, expressed as %.

6 October 2012 PROSPECTIVE ASSESSMENT OF NAFLD IN OBESITY 1167 Table 4. AUROCs of Several Scoring Systems for the Diagnosis of Significant Fibrosis and Advanced Fibrosis for the Design and Validation Cohorts Significant fibrosis Advanced fibrosis AUROC 1 AUROC 2 AUROC 1 AUROC 2 Antwerp NAFLD significant fibrosis score Antwerp NAFLD advanced fibrosis score NAFLD liver fat score FLI BARD score APRI FIB-4 index NAFLD fibrosis score Forns index AUROC, area under the receiver operating curve; BARD, body mass index, AST/ALT ratio, and diabetes mellitus; FLI, fatty liver index. gressive liver disease is mostly confined to NASH), the true prognostic significance of the different NAFLD lesions remains unclear. 1,10 We therefore examined NASH presence and severity and fibrosis as separate issues. It should be emphasized that the assessment of these different features can only be used to make an actual diagnosis, but until more data on NAFLD natural history become available, it cannot claim to be of prognostic value. In this population of apparently healthy people viewing themselves as simply overweight, the prevalence of significant liver disease was high. By using the definition of Brunt et al, 18 52% had NASH, the prevalence of which is generally reported to be 2% 5% in the Western population. 1 Almost 20% had significant fibrosis. These data not only stress the potential impact of NAFLD on health care but also the importance of screening populations at risk. We used two of the most common and largely overlapping definitions of NASH. 17,18 The scoring system by Kleiner et al, 17 designed for clinical research, is partially based on the score of Brunt et al 18 but adds the concept of the NAS to separately assess different features of NASH, to define NASH, and to score its severity. Recent data showed that although most patients have concordant results, a small proportion of patients are differently classified by both systems. 19 We choose to use both scoring systems, although the factors predicting NASH appeared to be grossly the same. The presence of NASH and its severity can reliably be assessed by combining anthropometry, biochemistry, and ultrasound. Our study confirms overweight, especially visceral fat accumulation, to be a risk factor for NASH. 1 Not body mass index but abdominal fat is predictive for NASH. Although subject to intraobserver and interobserver variation, waist circumference appears to be the strongest anthropometric predictive parameter. Computed tomography measured values, a more objective tool, 10,22 do not add to the diagnostic accuracy of the score. Elevated aminotransferase levels reflect liver damage. 23 Although they can be normal, they correlate quite well with NASH severity. 23 In our study, ALT was a stronger predictor than AST or the ALT/AST ratio. Not only the absolute value but also the cutoff of 40 U/L is highly predictive of NASH and therefore seems to represent an important diagnostic threshold. The formerly proposed lower cutoffs for hepatitis C are less accurate. 16 Insulin resistance has also been implicated in the pathogenesis of NASH, and this correlation is confirmed in our study. 2,24 C-peptide, reflecting pancreatic secretion, appeared to be the most reliable parameter. The combination of anthropometric parameters, aminotransferases, and parameters of insulin resistance is found in most of the reported scores. Interestingly, the NAFLD liver fat score, designed to diagnose liver fat and validated with magnetic resonance spectroscopy as the gold standard, has a rather high accuracy in the prediction of NASH. 3 The parameters included in this score largely cover the same factors as those used in our score. The other scoring systems had a much lower accuracy. They were, however, mostly designed to predict fibrosis and not NASH. The fatty liver index was designed to predict fatty liver but even so was not accurate for the diagnosis of NASH. Several attempts have been reported to increase the accuracy of noninvasive scores by adding more sophisticated tests. Our study clearly shows that computed tomography measured fat, ABT, -fetoprotein, serum cholinesterase, and other non-routine tests do not add to the accuracy of the scores and should not be used to screen for NASH. Even genetic testing did not increase the performance of the scores. PNPLA3 genotyping confirmed previous data on its significant association with NASH independently from metabolic factors but did not ameliorate the scoring. 21 This is in line with Kotronen et al, 3 who also found that PNPLA3 polymorphism did not increase the accuracy of the NAFLD liver fat score. In contrast, ultrasound, a simple, readily available, and cheap diagnostic tool, significantly increases the diagnostic accuracy of the prediction of NASH. Although inevitably subject to intraobserver and interobserver variation, it appears to be a powerful screening tool. Omitting ultrasound leads to low predictive values not meeting the criteria for clinical usefulness. Screening for NASH in patients with obesity should therefore include ultrasound. Only a few parameters appeared to be independent fibrosis predictors. Interestingly, AST but not ALT value was an important determinant, in contrast to what was observed for NASH. Although significant in univariate analysis, most of the other parameters did not reach significance in multivariate analysis. Ultrasound did not add to the diagnostic accuracy. FibroTest was not used, but some of the parameters included in the FibroTest were studied and failed to reach significance in multivariate analysis. 25 The role of CK-18 as a marker of fibrosis according to previous reports was confirmed. 20 However, CK-18 did not add to the accuracy of the scoring. PNPLA-3 polymorphism was not related to fibrosis, confirming recent data. 21 Although some are specifically designed to assess fibrosis, the overall accuracy of the classic scoring systems was low. This presumably results from the fact that they have been designed in selected patient groups with a high prevalence of more advanced disease. Another factor is the important role of diabetes in some scores. Although newly discovered diabetic patients were included in the analysis, the exclusion of patients already known to have diabetes (because of the potential impact on several metabolic and liver parameters) might have influ-

7 1168 FRANCQUE ET AL CLINICAL GASTROENTEROLOGY AND HEPATOLOGY Vol. 10, No. 10 enced the prevalence of fibrosis and hence the accuracy of the scores. From the published scores tested, the APRI score yields the highest accuracy. Because APRI consists of platelet count and AST, 6 its accuracy is presumably attributable to the predictive value of AST for fibrosis severity. Even our score has a rather unsatisfactory overall accuracy for the diagnosis of fibrosis. The negative predictive value is, however, very high. We propose to use the score to exclude the presence of advanced fibrosis. If the score does not exclude advanced fibrosis, a liver biopsy is still necessary to reliably assess liver fibrosis. Other studies are needed to investigate whether other diagnostic tools such as shear wave imaging are able to improve noninvasive scoring of liver fibrosis. 26 Supplementary Material Note: To access the supplementary materials accompanying this article, visit the online version of Clinical Gastroenterology and Hepatology at and at doi: / j.cgh References 1. Vernon G, Baranova A, Younossi ZM. Systematic review: the epidemiology and natural history of non-alcoholic fatty liver disease and non-alcoholic steatohepatitis in adults. Aliment Pharmacol Ther 2011;34: Vanni E, Bugianesi E, Kotronen A, et al. From the metabolic syndrome to NAFLD or vice versa? Dig Liver Dis 2010;42: Kotronen A, Peltonen M, Hakkarainen A, et al. Prediction of non-alcoholic fatty liver disease and liver fat using metabolic and genetic factors. Gastroenterology 2009;137: Bedogni G, Bellentani S, Miglioli L, et al. The fatty liver index: a simple and accurate predictor of hepatic steatosis in the general population. BMC Gastroenterol 2006;6: Harrison SA, Oliver D, Arnold HL, et al. Development and validation of a simple NAFLD clinical scoring system for identifying patients without advanced disease. Gut 2008;57: Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology 2003;38: Sterling RK, Lissen E, Clumeck N, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology 2006;43: Angulo P, Hui JM, Marchesini G, et al. The NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology 2007;45: Forns X, Ampurdanès S, Llovet JM, et al. Identification of chronic hepatitis C patients without hepatic fibrosis by a simple predictive model. Hepatology 2002;36: Musso G, Gambino R, Cassader M, et al. Meta-analysis: natural history of non-alcoholic fatty liver disease (NAFLD) and diagnostic accuracy of non-invasive tests for liver disease severity. Ann Med 2011;43: Francque S, Verrijken A, Mertens I, et al. Non-cirrhotic human NAFLD induces portal hypertension in relation to the histological degree of steatosis. Eur J Gastroenterol Hepatol 2010;22: Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults. Executive summary of the third report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III). JAMA 2001;285: Alberti KG, Zimmet P, Shaw J, et al. The metabolic syndrome: a new worldwide definition. Lancet 2005;366: Merkel C, Bolognesi M, Bellon S, et al. Aminopyrine breath test in the prognostic evaluation of patients with cirrhosis. Gut 1992; 33: Saverymuttu SH, Joseph AE, Maxwell JD. Ultrasound scanning in the detection of hepatic fibrosis and steatosis. Br Med J (Clin Res Ed) 1986;292: Prati D, Taioli E, Zanella A, et al. Updated definitions of healthy ranges for serum alanine aminotransferase levels. Ann Intern Med 2002;137: Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology 2005;41: Brunt EM, Janney CG, Di Bisceglie AM, et al. Nonalcoholic steatohepatitis: a proposal for grading and staging the histological lesions. Am J Gastroenterol 1999;94: Brunt EM, Kleiner DE, Wilson LA, et al. Nonalcoholic fatty liver disease (NAFLD) activity score and the histopathologic diagnosis in NAFLD: distinct clinicopathologic meanings. Hepatology 2011; 53: Younossi ZM, Jarrar M, Nugent C, et al. A novel diagnostic biomarker panel for obesity-related nonalcoholic steatohepatitis (NASH). Obes Surg 2008;18: Rotman Y, Koh C, Zmuda JM, et al. The association of genetic variability in patatin-like phospholipase domain-containing protein 3 (PNPLA3) with histological severity of nonalcoholic fatty liver disease. Hepatology 2010;52: van der Koov K, Seidell JC. Techniques for the measurement of visceral fat: a practical guide. Int J Obes Relat Metab Disord 1993;17: Suzuki A, Lymp J, Sauver JS, et al. Values and limitations of serum aminotransferases in clinical trials of nonalcoholic steatohepatitis. Liver Int 2006;26: Bugianesi E, McCullough AJ, Marchesini G. Insulin resistance: a metabolic pathway to chronic liver disease. Hepatology 2005;42: Lassailly G, Caiazzo R, Hollebecque A, et al. Validation of noninvasive biomarkers (FibroTest, SteatoTest, and NashTest) for prediction of liver injury in patients with morbid obesity. Eur J Gastroenterol Hepatol 2011;23: Kim KM, Choi WB, Park SH, et al. Diagnosis of hepatic steatosis and fibrosis by transient elastography in asymptomatic healthy individuals: a prospective study of living related potential liver donors. J Gastroenterol 2007;42: Reprint requests Address requests for reprints to: Sven Francque, MD, PhD, Department of Gastroenterology and Hepatology, Antwerp University Hospital, Wilrijkstraat 10, B-2650 Edegem, Belgium. sven.francque@ uza.be; fax: Conflicts of interest The authors disclose no conflicts. Funding This work is part of the project Hepatic and Adipose Tissue and Functions in the Metabolic Syndrome (HEPADIP), which is supported by the European Commission as an Integrated Project under the 6th Framework Program (contract LSHM-CT ).